Synaptic rescue in an Alzheimer’s mouse model: low-temperature steam-derived black ginseng oligosaccharides remodel protein S-nitrosylation-NADPH oxidase axis

Abstract

Synaptic loss and aberrant protein S-nitrosylation (SNO) are strongly linked to cognitive decline in both patients and animal models of Alzheimer’s disease (AD). Our recent work in an AD mouse model has shed light on the role of oligosaccharides extracted from black ginseng (OSBG) in ameliorating cognitive impairment. However, the precise molecular mechanisms responsible for therapeutic efficacy of OSBG in AD are not well understood. In the present study, we employed an innovative SNOTRAP-based proteomic approach to quantify SNO proteins in the brain of APP/PS1 mice following OSBG intervention. The results revealed that differentially expressed SNO proteins, such as SNO-NOX1 and SNO- NOX5, confirmed by Western blot (WB), are significantly enriched in pathways related to oxidative stress, such as “Oxidative_stress activation of NADPH oxidase” and “Synaptic target recognition”. OSBG treatment significantly alleviated oxidative stress via inhibition of NADPH oxidase activity in APP/PS1 mouse and PC12 cells by WB immunofluorescence (IF) assays. More importantly, upregulation of PSD-95 and SYN1 was detected in the hippocampal tissue of APP/PS1 mice after OSBG intervention, which was further validated by the corresponding mRNA expression levels. Consistently, histopathological analysis revealed the restoration of hippocampal cellular structure. Overall, our findings highlight the synapse-protective effect of OSBG in an AD model through regulating protein SNO levels and inhibiting NADPH oxidase activity, revealing a novel mechanism by which OSBG alleviated oxidative stress injury.

Introduction

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder primarily featured by cognitive decline and behavioral disturbances^1,2. Since its initial clinical description in 1907, AD has reached epidemic proportions. As of 2023, an estimated 6.7 million Americans aged 65 years and older are affected, with this figure projected to increase to 13.8 million by 2060^3,4. Despite over a century of research efforts, the precise pathogenesis of AD remains incompletely understood. The leading pathophysiological hypotheses mainly involve: (1) the amyloid plaque hypothesis, (2) neurofibrillary tangle formation, and (3) progressive neuronal loss. Emerging evidence implicates multiple synergistic mechanisms in AD progression^5–10. Although the amyloid hypothesis holds certain rationality, it still has significant limitations, especially since the association between amyloid production and the development of neurofibrillary tangles has not been fully elucidated. Therefore, there is an urgent need to establish a comprehensive understanding of neurodegenerative processes and the mechanisms underlying neuronal death. Free radicals are highly likely to play a crucial role in the aforementioned pathological processes. Based on the free radical theory of aging proposed decades ago, this hypothesis posits that the accumulation of reactive oxygen species (ROS) with advancing age induces damage to key cellular components, including the cell nucleus, mitochondrial DNA (mtDNA), cell membrane, and cytoplasmic proteins. Such an imbalance between free radical production and ROS scavenging is regarded as a key driving factor in the pathogenesis of most neurodegenerative diseases, including AD, which has garnered sustained support in the academic community^11,12. Multiple studies have confirmed that oxidative stress products exert distinct toxic effects on specific cellular targets in AD patients. Specifically, significant oxidative damage is observed in mtDNA in the parietal cortex of AD patients, while the oxidative level of nuclear DNA is relatively low¹³. Protein oxidation is found in both AD patients and non-AD elderly individuals, but the level of protein oxidation in AD patients is significantly higher than that in non-AD elderly individuals in brain regions with the most prominent AD pathological changes¹⁴. Meanwhile, multiple independent studies have detected abnormal accumulation of a variety of peroxidation end products in the brain tissues of AD patients, and this phenomenon is particularly prominent in regions enriched with neurofibrillary tangles, including malondialdehyde¹⁵, peroxynitrite¹⁶, carbonyl compounds¹⁷, advanced glycation end products¹⁸, superoxide dismutase 1 (SOD-1)¹⁹, and heme oxygenase 1 (HO-1)²⁰. The above series of evidence fully indicates that there is a direct correlation between the level of oxidative stress in the brain and the occurrence and development of AD. Currently, only six FDA-approved medications exist for AD management, predominantly ascribed to acetylcholinesterase inhibitors and N-methyl-D-aspartate (NMDA) receptor antagonists^21,22. As research into the pathological mechanisms of AD has deepened, immunotherapies targeting its core pathological pathways have emerged prominently in recent years. Within the spectrum of immunotherapeutic agents, a primary strategy involves targeting amyloid-β (Aβ) for plaque clearance. However, most immunotherapeutic agents targeting Aβ are still in clinical trials. These pharmacological interventions demonstrate limited disease-modifying capabilities, largely restricted to transient symptomatic relief with modest clinical efficacy. Furthermore, long-term pharmacotherapy may lead to unintended side effects, for instance, headache and amyloid-related imaging abnormalities³. This critical unmet need underscores the urgency for developing novel therapeutics that combine potent neuroprotective properties with enhanced safety profiles.

Traditional Chinese medicine (TCM) represents an attractive therapeutic alternative for AD, offering multi-target modulation capabilities combined with favorable safety profiles²³. The root of Panax ginseng C.A. Meyer has been employed as both a functional dietary ingredient and a health-enhancing agent in traditional medicine for millennia. Phytochemical analyzes has revealed that P. ginseng contains various pharmacologically active constituents, such as oligosaccharides, saponins, and amino acids²⁴. Accumulating preclinical evidence has demonstrated that ginseng oligosaccharides exhibit a range of biological activities, comprising antioxidant effects²⁵, immunomodulatory properties²⁶, and neuroprotective functions in central nervous system^27. Black ginseng (BG) represents a unique category of processed ginseng, derived from white ginseng through a distinctive process of repeated steaming and drying until its transformation into a black coloration, resulting in significant phytochemical transformation²⁶. Mass spectrometry (MS) analysis has revealed that oligosaccharides from black ginseng (OSBG), innovatively prepared via a low-temperature steam-heating process in our group, possess three novel components absent in other ginseng varieties^28,29. Such a profound compositional distinction strongly implicates OSBG in possessing augmented therapeutic potential, especially for addressing the multifactorial neurodegenerative processes inherent in AD. Recently, our group has innovatively prepared OSBG using a low-temperature steam-heating process, which has shown promising effects in ameliorating cognitive impairment associated with AD by activating the Keap-1/Nrf2 pathway³⁰.

Nitric oxide (NO) that is a crucial signaling molecule plays a key role in synaptic transmission and neuronal function^31,32. NO can influence ion transporters and S-nitrosylation (SNO) of proteins^32–34. SNO is a post-translational modification of proteins where NO is covalently attached to the thiol group of a cysteine residue, thereby affecting the function of the protein and downstream signal transduction³⁵ Under normal physiological conditions, SNO is involved in a number of processes, consisting of protein localization, axonal transport, maintenance of synaptic plasticity, and regulation of various neuronal pathways^34,36. Simultaneously, dysregulated SNO levels are associated with increased oxidative stress in the context of AD^37–39.

To the best of our knowledge, this is the first report on the effect of oligosaccharides on SNO levels of protein in brain tissue. Here, we used SNO proteomics based on SNOTRAP probe (a triaryl phosphine derivative) technology to determine SNO proteins in the whole brain tissues of APP/PS1 mice, followed by investigation of biological processes and signaling pathways enriched by SNO proteins. We focused on the oxidative stress and synaptic target recognition pathways associated with AD. NADPH oxidase and synaptic markers were subsequently examined in the whole brain tissues of APP/PS1 mice after OSBG intervention. Our research holds great significance in uncovering the mechanisms of OSBG in AD and could pave the way for new therapeutic strategies.

Results

The composition of the water extract

The monosaccharide residues in OSBG were identified as mannose, rhamnose, glucose, glucuronic acid, galactose, arabinose, and fructose in our previous report³⁰. Besides, eight carbohydrates—glucose, fructose, sucrose, maltose, isomaltose, maltotriose, nystose, and maltotetraose were involved in BG³⁰. In this study, molecular weight and homogeneity of OSBG were determined by high-performance gel permeation chromatography. The analysis revealed that OSBG exhibited a number-average molecular weight (Mn) of 2.272 kDa, a weight-average molecular weight (Mw) of 2.961 kDa, and a peak molecular weight (Mp) of 1.371 kDa (Fig. S1, and Supplementary Data 1). The polydispersity index, calculated as Mw/Mn, was 1.303, indicating a relatively narrow molecular weight distribution.

According to the editor’s suggestion, we do detect peptides and glycopeptides by thin-layer chromatography and Fourier transform-infrared (FT-IR) spectroscopy (Figs. S2 and S3, and Supplementary Data 1). The amino acid sequence of the peptides was then determined by mass spectrometry (MS), indicating the presence of MDVRRRSVTKTLTAGEPLKSQNQHSSS, MTLNAQQVVLITGAGSGIGKAIALALASAARTLSGSLHCEQGSR, MTRTIAAMLLVVTAALAGCNTVAGMGQDISKGGQAISDTAEKAK, TTVFSHSQTVVVCGNCHAVL, and YVLPYGQMSLWG (Supplementary Data 2). All the corresponding experiments and descriptions were provided in the Supplementary materials. In summary, OSBG comprised not only oligosaccharides but also glycopeptide and peptide.

S-nitrosoproteomic analysis in the whole brain tissues of the control group, APP/PS1 group, and APP/PS1+OSBG group

SNO proteins and their SNO sites in the mouse whole brain tissues were identified using the SNOTRAP proteomics strategy. SNO proteins identified in all three groups were detailed in Supplementary Data 3 and 4 as well as Fig. 1A. Specifically, identified SNO proteins and sites were: 687 proteins and 1028 sites (control group), 656 proteins and 977 sites (APP/PS1 group), and 673 proteins and 998 sites (APP/PS1+OSBG group). This result provided a large database for SNO proteins and SNO sites in the mouse whole brain tissues (Supplementary Data 5). The data indicated that approximately 85% of SNO proteins were common between the two or three tissue groups (Fig. 1B), thereby demonstrating the presence of differential SNO proteins after OSBG treatment. Among them, 172 SNO proteins were significantly upregulated in the APP/PS1 group, whereas 209 SNO proteins were significantly downregulated after OSBG intervention (Supplementary Data 6). Compared with the control group, only 8% of SNO proteins were downregulated, while 23% were upregulated in the APP/PS1 group (Fig. 1C). In contrast, compared with the APP/PS1 group, 28% of SNO proteins were downregulated and 12% were upregulated in the APP/PS1+OSBG group (Fig. 1D). These differentially regulated SNO proteins demonstrate that OSBG can significantly reverse the abnormal modification of some key differentially regulated SNO proteins, providing core targets for subsequent pathway enrichment analysis and functional validation of these proteins.

Fig. 1. S-nitrosoproteomic analysis in the whole brain tissues of the control group, APP/PS1 group, and APP/PS1+OSBG group.

A The histogram illustrates the number of SNO proteins identified in whole brain tissues across different mouse cohorts. B Venn diagrams were utilized to visualize the quantitative distribution of SNO proteins detected in whole brain tissue samples in the three groups of mice. C Volcano plot showing the distribution of upregulated and downregulated SNO proteins in the whole brain tissues of the APP/PS1 group compared with the control group. D Volcano plot showing the distribution of upregulated and downregulated SNO proteins in the whole brain tissues of the APP/PS1 + OSBG group compared with the APP/PS1 group. The data are presented as the mean ± SEM (n = 4).

GO and KEGG pathway analysis of the SNO-proteome

In order to gain a systematic insight into the function of SNO-proteins, the GO processes and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway functions were analyzed using differential proteins belonging to the Control, APP/PS1, and APP/PS1+OSBG groups (detailed lists in Supplementary Data 7). The biological processes and signaling pathways in the control group exhibited a notable enrichment in fundamental pathways associated with cell growth and metabolic processes (Figs. 2A and 3A), including “Positive regulation of cell differentiation”, and “Cell adhesion_Cell-matrix glycoconjugates”. GO processes and KEGG pathways related to AD, such as “Nervous system development”, “Neurophysiological process_Activity-dependent synaptic AMPA receptor removal”, “Oxidative stress_Activation of NOX1, NOX5, DUOX1 and DUOX2 NADPH Oxidases”, and “Modulation of chemical synaptic transmission”, were markedly enriched in the APP/PS1 group (Figs. 2B and 3B). The APP/PS1 + OSBG group exhibited a prominent enrichment in the functions of “Neurophysiological process_Synaptic vesicle fusion and recycling in nerve terminals”, “Oxidative stress_Activation of NOX1, NOX5, DUOX1 and DUOX2 NADPH Oxidases”, “Regulation of synapse structure or activity”, “Oxidative stress_Activation of NADPH oxidase”, and other processes that are also associated with AD (Fig. 2C and 3C). It was noteworthy that SNO-synaptophysin and SNO-pan-PKC (PKC-beta isoform), both downregulated by OSBG treatment (Supplementary Data 6), were functionally involved in the first two pathways. In short, these biological processes and pathway analyzes offer insights into the potential regulation of SNO-mediated proteins during AD pathogenesis.

Fig. 2. GO analysis was conducted on the differential SNO proteins.

A Control group, B APP/PS1 group, C APP/PS1 + OSBG group (showing the results of GO analysis for each group, respectively).

Fig. 3. KEGG pathway analysis was conducted on the differential SNO proteins.

A Control group, B APP/PS1 group, C APP/PS1 + OSBG group (showing the results of KEGG pathway analysis for each group, respectively).

Aside from functional clustering analyzes, we used String analysis to explore possible interactions among the downregulated SNO proteins in the APP/PS1 + OSBG group (Supplementary Data 8), where 115 SNO proteins passed our cutoff criteria and were classified as interactors. The downregulated SNO proteins following OSBG intervention were significantly enriched in pathways including “respiratory electron transport”, “pyruvate metabolism and tricarboxylic acid (TCA) cycle”, “RhoGTPase-activated Nicotinamide Adenine Dinucleotide Phosphate (NADPH) oxidase”, and “GABAB receptor activation” (Fig. 4). These pathways are associated with NADPH oxidase, energy metabolism, and oxidative stress, further underscoring the potential link between the OSBG-induced downregulated SNO proteins and AD pathogenesis.

Fig. 4. Interaction network analysis conducted on the set of down-regulated SNO proteins after OSBG intervention.

Regulated functional protein clusters are displayed by shadows.

OSBG attenuated H₂O₂-induced oxidative stress in PC12 cells

To further confirm the antioxidant activity of OSBG, cell viability assays were performed, achieving an IC₅₀ value of approximately 80 μM for H₂O₂ (Fig. 5A). Subsequently, the effect of OSBG on cell proliferation was evaluated by measuring cell viability. OSBG exerted a significant inhibitory effect on PC12 cell viability at concentrations exceeding 200 μg/mL, with a time-dependent reduction in cell viability observed (Fig. 5B–E). Consequently, subsequent viability assays were focused on concentrations below this threshold. After H₂O₂ induction, cell viability was significantly improved in the groups treated with 25 μg/mL OSBG (Fig. 5F), which was ultimately selected for subsequent experiments. To further investigate the effect of OSBG on intracellular ROS and confirm its antioxidant activity, we detected ROS levels in PC12 cells. Compared with the control group, the ROS fluorescence intensity was significantly increased in the phorbol 12-myristate 13-acetate (PMA, NADPH oxidase activator) and model groups, indicating the activation of intracellular oxidative stress (Fig. 5G, H). Following OSBG treatment, intracellular ROS levels were significantly reduced compared with the PMA and model groups (^****p < 0.001), demonstrating that OSBG can reverse PMA- and H₂O₂-mediated oxidative stress (Fig. 5G, H). In addition, we measured the intracellular levels of 3-nitrotyrosine (3-NT) and 4-hydroxynonenal (4-HNE), which are validated biomarkers of oxidative stress. Compared with the PMA group and the model group, OSBG intervention alleviated intracellular oxidative stress (Fig. 5I, J).

Fig. 5. OSBG attenuated H₂O₂-induced oxidative stress in PC12 cells.

A Cell viability values of PC12 cells induced by different concentrations of H₂O₂ (1–450 μM). Data are presented as the mean ± standard error of the mean (SEM) (n = 6). B–E Cell viability of PC12 cells treated with various concentrations of OSBG revealed no significant effect on cell survival when OSBG was below 100 μg/mL. F Using concentrations of OSBG with minimal effects on cell viability (≤100 μg/mL), cell viability was assessed in the control, model, and OSBG-treated groups, and 25 μg/mL OSBG exhibited significant antioxidant activity. G Cells were stained with the ROS detection kit and imaged under a fluorescence microscope. H Relative fluorescence intensity of PC12 cells. Six intact cells per group were selected for relative fluorescence intensity analysis. I 3-NT expression levels in each group. J 4-HNE expression levels in each group. K Individual protein bands and corresponding group labels. L Protein expression levels of p47^phox and P-p47^phox. Data are presented as mean ± standard error of the mean (SEM) (n = 6). ^**p < 0.01, ^****p < 0.0001 compared with the model group; ^##p < 0.01, ^###p < 0.001, ^####p < 0.0001 compared with the PMA group.

To validate that OSBG acts via the NADPH oxidase-p67^phox/PKC/p47^phox signaling, rescue assays were performed. WB was used to assess p47^phox (also known as neutrophil cytosolic factor 1) and P-p47^phox (phosphorylated p47^phox), a key activator of the NADPH oxidase complex^40–42. The results clearly indicated that OSBG intervention led to a downregulation of p47^phox expression (Fig. 5K, L). Briefly, OSBG exerted an inhibitory effect on the activation of NADPH oxidase in PC12 cells triggered by either PMA or H₂O₂.

OSBG reduced oxidative stress in APP/PS1 mouse brains

To confirm the robustness of the enriched pathway “NADPH oxidase” after OSBG intervention by SNOTRAP with MS analysis, we examined the expression of several key proteins related to it (Fig. 6A). In comparison with the control group, the APP/PS1 group revealed a significant increase in the expression of pan-PKC protein (^**p < 0.01). OSBG was observed to significantly downregulate the expression of pan-PKC protein to varying degrees (^*p < 0.05) (Fig. 6B). Additionally, in contrast to the model group, OSBG could significantly downregulate the levels of p67^phox (Fig. 6C). As for the expression of RAC2, there is no remarkable influence after OSBG intervention (Fig. 6D). The findings suggest that OSBG mitigates oxidative stress in the brains of APP/PS1 mice by modulating the expression of multiple pivotal proteins associated with NADPH oxidase activation.

Fig. 6. OSBG reduced oxidative stress in whole brain tissues of APP/PS1 mice.

A–D Immunoblotting images and quantification of pan-PKC, p67^phox, and RAC2 in whole brain tissues of APP/PS1 mice. Band intensities were measured using ImageJ and analyzed using one-way ANOVA. E, F Biotin-switch assay and WB analysis of SNO-NOX1 and SNO-NOX5 in whole brain tissues from APP/PS1 mice administered with OSBG (40 mg/kg/day) or normal saline (10 mL/kg/day) for 8 weeks. G–I Effect of OSBG on SOD, MDA, and NO levels in APP/PS1 mouse brain. The data are presented as the mean ± SEM (n = 3). For all the analysis, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, ^****p < 0.0001, ns: no significance versus APP/PS1 group.

KEGG pathway enrichment analysis revealed that the “Oxidative stress_Activation of NOX1, NOX5, DUOX1 and DUOX2 NADPH Oxidases” pathway was one of the major pathways enriched in the control group (Fig. 3A). Subsequently, compared with whole brain tissue samples from the control group, significantly increased levels of SNO-NOX1 and SNO-NOX5 were observed in the APP/PS1 group (^***p < 0.001 and ^****p < 0.0001). The OSBG treatment group significantly reversed the elevated SNO-NOX1 and SNO-NOX5 levels (Fig. 6E, F) (^*p < 0.05 and ^***p < 0.001).

Additionally, the expression levels of antioxidant enzymes in the brains of APP/PS1 mice were also examined. The content of SOD in the whole brain tissues of APP/PS1 mice was found to be significantly reduced in comparison to the control group (^**p < 0.01), while the levels of MDA and NO were observed to be significantly elevated (^***p < 0.001) (Fig. 6G–I). The administration of OSBG to APP/PS1 mice has been demonstrated to markedly elevate the level of SOD (^*p < 0.05), while concurrently reducing the contents of MDA (^***p < 0.001) and NO (^**p < 0.01) (Fig. 6G–I). These findings further suggest that OSBG can reduce oxidative stress-related injury in the brains of AD mice.

OSBG improved synaptic plasticity in the brains of APP/PS1 mice

nNOS represents one of the subtypes of NO synthase (NOS). It is primarily expressed in specific neurons within the central nervous system, where it modulates synaptic plasticity and neurogenesis⁴³. The expression level of nNOS protein in whole hippocampal tissue was detected by WB (Fig. 7A, B). In contrast, the level of nNOS in the hippocampal tissue of APP/PS1 mice was markedly reduced following OSBG intervention (^***p < 0.001), which aligns with the observed trend in the control group. Next, the expression levels of SYN1 and PSD-95, which are used as biomarkers of synapses, were determined through WB, qRT-PCR and IF to indirectly reflect synaptic plasticity. In comparison to the APP/PS1 group, the expression levels of PSD-95 and SYN1 were markedly elevated following OSBG intervention (^*p < 0.05 and ^****p < 0.0001) (Fig. 7A, B). This phenomenon was also corroborated by the results from the IF (Fig. 7C, D) and qRT-PCR experiments (Fig. 7E). OSBG intervention significantly increased the fluorescence intensity of PSD-95 (^**p < 0.01) and SYN1 (^*p < 0.05) (Fig. 7C, D). Consistent with the above findings, the qRT-PCR results demonstrated the same trend.

Fig. 7. OSBG improved synaptic plasticity in APP/PS1 mouse brain.

A, B Immunoblotting images and quantification of nNOS, PSD-95, and SYN1 in hippocampal tissues of APP/PS1 mice. n = 3 for each group. C, D Immunostaining images and relative immunofluorescence intensities of PSD-95 and SYN1 in hippocampal tissue sections of APP/PS1 mice. Scale bars, 50 μm. n = 6 for each group. E The mRNA expression of PSD-95 and SYN1 in the hippocampal tissues of APP/PS1 mice using qRT-PCR analysis. n = 6 for each group. The data are presented as the mean ± SEM. Band and fluorescence intensities were measured using ImageJ and analyzed using one-way ANOVA. For all analyses, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, and ^****p < 0.0001 versus APP/PS1 group.

These results suggest that OSBG influences synaptic function and protein expression by affecting the overall process of synaptic plasticity protein synthesis. Consequently, OSBG may represent a promising therapeutic avenue for addressing the progression of AD.

OSBG attenuated hippocampal tissue damage in APP/PS1 Mice

HE staining was performed to evaluate tissue damage, where the size, density, arrangement and staining of neuronal cells in hippocampal tissue served as indications of neuronal damage⁴⁴. Considering the hippocampal region in the brain playing a central role in learning and memory processes, we specifically performed HE staining analysis on this region. Histological examination revealed tightly arranged hippocampal tissue cells with no observable loss (black arrows), while neurons exhibited intact morphology with clearly defined nuclear membranes and nucleoli (white arrow) (Fig. 8A). The hippocampal tissue cells with significant neuronal loss (black arrows) were scattered in APP/PS1 mice, accompanied by perineuronal edema, cellular atrophy, and indistinct intercellular boundaries (white arrows) (Fig. 8B). Following OSBG treatment, the pathological condition of certain neurons in the hippocampal tissue showed improvement (black arrows), with clearer delineation of cell boundaries and nuclei (white arrows) (Fig. 8C). The findings strongly suggest that OSBG intervention is protective against hippocampal tissue damage.

Fig. 8. OSBG alleviated hippocampal tissue lesions in APP/PS1 mice (HE staining).

A Control group. B APP/PS1 group. C APP/PS1 + OSBG group. Scale bar: 50 μm.

Discussion

Given the multifactorial complexity of AD pathogenesis, TCM characterized by its multi-target pharmacological profile presents a promising avenue for developing novel AD therapeutics⁴⁵. Our prior research demonstrated that OSBG attenuated scopolamine-induced memory deficits by modulating the gut microbiota⁴⁵. This study integrates innovative SNOTRAP-MS proteomic and biochemical analyzes to elucidate the mechanism of OSBG-mediated neuroprotection, and demonstrates the involvement of NO. The results suggest the involvement of NADPH oxidase, NO, and synapse in the mechanism of cognitive behavior alteration in OSBG-treated APP/PS1 mice. Beyond its role in synaptic transmission within the brain⁴⁶, NO functions prominently in signaling pathways through SNO of peptides and proteins^47,48. SNO, a post-translational modification, regulates the localization and activity of key enzymes and receptors^49,50, ultimately modulating oxidative stress signaling pathways, synaptic plasticity, and protein assembly^41,51,52. SNO of oxidative stress related proteins in the APP/PS1 group may exacerbate oxidative stress processes in whole brain tissues and affect the dynamic balance of redox, which is one of the most common phenomena in neurological disorders⁵³.

OSBG treatment in APP/PS1 mice reduced SNO levels in key components of NADPH oxidase, including NOX1, NOX5, DUOX1, and DUOX2 (Figs. 2 and 3), which act as critical drivers of neuronal damage^54–56. Consistent with this, decreased levels of SNO-NOX1 and SNO-NOX5 were detected in the brains of APP/PS1 mice after OSBG intervention (Fig. 6E, F), revealing the regulatory role of S-nitrosylation in NOX-mediated oxidative stress in AD. Notably, this study represents a pioneering investigation into the S-nitrosylation of NOX1/NOX5 and OSBG, revealing their presence and potential contribution in the APP/PS1 mouse model. The significant inhibition of the PMA/H₂O₂-induced ROS burst by OSBG in PC12 cells (Fig. 5G, H) is consistent with the established role of the NOX family as a primary source of ROS. This result revealed that OSBG reduced oxidative stress both in vivo and in vitro (Figs. 6G–I and 5I) by interfering with NOX activity.

The activation of NADPH oxidase is dependent on the cytosolic regulatory subunits such as p47^phox and p67^phox57. Upon stimulation by external signals like bacterial lipopolysaccharides or chemokines, the intracellular protein kinase PKC is activated and subsequently phosphorylates p47^phox58. The phosphorylated p47^phox can associate with p67^phox to form a cytosolic complex, which then translocates to the cell membrane. At the membrane, this complex interacts with activated Rac GTPase, leading to the assembly of the fully functional NADPH oxidase complex⁵⁹. Our results demonstrated that OSBG suppressed PKC activation (Fig. 6A, B), decreased p67^phox expression (Fig. 6A, C), and blocked the phosphorylation of p47^phox (Fig. 5J, K), which led to effective suppression of NADPH oxidase activity and a reduction in ROS production in Fig. 5, revealing a novel mechanism by which OSBG alleviates oxidative stress injury.

Furthermore, GO and KEGG analyzes indicated the enrichment of biological processes involved in synaptic target recognition. Given the established link between synaptic morphological function and various neurological diseases, including PD and AD^60,61, these findings are significant. Abnormalities in synaptic recognition and function are key pathological features of AD⁵⁰. A postsynaptic protein involved in synaptic recognition and plasticity, PSD-95, plays a critical role in linking NMDA receptor activity to NO production and subsequent neurotoxicity^62,63. It has been reported that PSD-95 activity is inhibited by S-nitrosylation, which affects synaptic recognition and aggregation⁶⁴. Meanwhile, significantly reduced activity of this protein has been reported in patients carrying the APOE4 gene, the most common major genetic risk factor for late-onset AD⁶⁵. Of particular interest, OSBG treatment significantly altered PSD-95 expression (Fig. 7). Building upon prior observations of dysregulated S-nitrosylation of SYN1 in aging rodents⁶⁶, we found that OSBG administration not only decreased SNO modification of synaptophysin (Supplementary Data 6) but also remarkably enhanced SYN1 protein abundance, as validated by triplicate methodological approaches (Fig. 7). We propose that regulation of SNO levels of synaptic recognition proteins may underlie the treatment of learned memory deficits in AD.

We suggest that OSBG could regulate imbalances in NADPH oxidase and synaptic homeostasis by altering SNO. Dysfunction of critical proteins caused by aberrant SNO results in impairment of neurodevelopmental processes, neuronal function, and neuronal cell viability, which are also features observed in AD pathology. In several studies, NADPH inhibitors are found to be protective in neurodegenerative disorders such as PD^67,68, AD^69,70 and in other toxic injuries⁶⁰. Our findings suggest that OSBG, a potential NADPH inhibitor, may alleviate hippocampal tissue damage, as demonstrated by HE staining results in Fig. 8, supporting its use as a therapeutic strategy for AD. However, considering water extracts from black ginseng include other compounds (such as peptides and glycopeptides) in addition to oligosaccharides, a limitation of this research lies in its scope, which primarily focused on oligosaccharides without comprehensively investigating the potential contributions of peptides and glycopeptides. This limitation must be considered when interpreting the results from the water extracts, as they underscore potential constraints inherent in the extraction method. Grasping this limitation is vital for a thorough and precise understanding of the roles of water extracts from black ginseng.

Methods

Reagents

Protease cocktail inhibitor, 4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), acetic acid, iodoacetamide, acetonitrile (ACN), formic acid (FA), Triton X-100, and ethylenediaminetetraacetic acid (EDTA) were bought from Sigma (USA). Bovine serum albumin (BSA) digest standard, N-ethylmaleimide (NEM), peptide desalting spin columns, and high-capacity streptavidin agarose resin were sourced from Pierce (Thermo Fisher, Rockford, USA). Vivaspin Turbo 4 filters with a nominal molecular-weight cutoff of 10 kD were provided by Fisher Scientific (MA, USA). Sodium dodecyl sulfate (SDS) and tris (2-carboxyethyl) phosphine (TCEP) were purchased from Aladdin Biochemical Technology Co., Ltd (Shanghai, China). SOD assay kit (ml076328) and malondialdehyde (MDA) (ml093018) assay kit were purchased from Shanghai Enzyme-linked Biotechnology Co., Ltd (Shanghai, China). NO assay kit (A013-2-1) was purchased from Nanjing Jiancheng Biotechnology Research Institute Co., Ltd (Nanjing, China). Sequencing grade modified trypsin was obtained from Promega (Madison, WI). Antibodies against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (10494-1-AP), synaptophysin 1 (SYN1) (20258-1-AP), postsynaptic density protein 95 (PSD-95) (30255-1-AP), protein kinase C, beta (pan-PKC) (12919-1-AP), ras-related C3 botulinum toxin substrate 2 (RAC2) (60077-1-Ig), and p67^phox (15551-1-AP) were purchased from Proteintech Group, Inc. (Wuhan, China). Neuronal NOS (nNOS) (#4231) was sourced from Cell Signaling Technology (USA). Goat anti-rabbit IgG H&L antibody (bs-0295G-HRP) was purchased from Beijing Bioss Biotechnology Co., Ltd (Beijing, China).

Preparation of OSBG

BG was prepared by incubating 5-year-old fresh ginseng, sourced from the Wanliang market (Jilin, China), at 70 °C for 15 days. BG oligosaccharides were extracted by water and filtered using previously established methods⁷¹. The oligosaccharide sample was re-dissolved in water and dialyzed using an ultrafiltration membrane with a molecular weight cut-off of 3000 Da. The dialysate was collected and lyophilized to obtain the OSBG sample.

Animals and drug treatment

A total of 20 APP/PS1 mice (3 months old, Nanjing Shuboen Biotechnology Co., Ltd) were randomly distributed into two groups: APP/PS1 group and APP/PS1+OSBG group. 10 wild-type C57BL/6N mice, 3 months of age, were obtained from Liaoning Changsheng Biotechnology Co., Ltd (Shenyang, China) and designated as the control group. The study was reviewed and approved by the Ethics Committee, Changchun University of Chinese Medicine (2025041). The mice in each group were approximately half male and half female. The mice were allowed to acclimatise for a period of a week prior to the commencement of the experiments. During this time, they were housed in the animal room at a temperature of 22–25 °C, provided with food and water ad libitum.

The mice in the control and APP/PS1 groups were administered 10 mL/kg saline via gavage on a daily basis, while OSBG at a dose of 40 mg/kg was administered daily to the mice in the APP/PS1+OSBG group via gavage³⁰. The experimental design outlined in Table 1 was strictly adhered to. The mice in each group were administered the relevant treatment continuously for a period of 8 weeks. After the completion of drug administration, the mice were euthanized by means of deep isoflurane anesthesia combined with cervical dislocation. Deep isoflurane anesthesia ensured that the mice remained completely unconscious and pain-free during cervical dislocation. Specifically, the mice were exposed to 3.0% isoflurane (oxygen flow rate: 1.0–1.5 L/min) for 3–5 min to induce deep anesthesia, until the loss of spontaneous movement, disappearance of the corneal reflex and pedal withdrawal reflex were observed. Subsequently, an experienced operator firmly grasped the mouse’s head with the thumb and index finger and pulled the base of the tail gently yet firmly in the opposite direction to transect the spinal cord, leading to the immediate cessation of cardiac and respiratory functions. The death of the mice was confirmed by two definitive criteria: the absence of thoracic movement and full pupillary dilation with no response to light stimulation. Finally, the whole brain tissues and hippocampal tissues of the mice were harvested and stored at −80 °C for subsequent analysis.

Table 1.

The administration regime and dosage in the three groups

Group	Species	Experiment treatment (8 weeks)
Control group	C57BL/6	saline solution (10 mL/kg/d)
APP/PS1 group	APP/PS1	saline solution (10 mL/kg/d)
APP/PS1 + OSBG group	APP/PS1	OSBG (40 mg/kg/d)

Whole brain tissue homogenization and sample preparation for MS

The preparation of whole brain tissues was performed using our previously described method^48,72. In brief, whole brain tissues were homogenized into lysis buffer HEPES, followed by centrifuging the homogenates at 16,000 × g for 20 min at 4 °C. The lysates were centrifuged with the spin filters after being washed once with HEPES. SNO was selectively converted to steady disulfide-iminophosphorane by reaction with 1.5 mM SNOTRAP solution, using 10 mg of initial protein, followed by trypsin digest. Next, each digested sample was incubated with high-capacity streptavidin agarose beads. Subsequently, the bound SNO peptides were eluted with TCEP and alkylated with NEM. Samples were then desalted prior to MS analysis.

NanoLC-MS analysis

NanoLC-MS/MS analysis was performed using an Orbitrap Exploris 480 mass spectrometer coupled to a Vanquish™ Neo ultrahigh-performance liquid chromatography (UHPLC) system (Thermo Scientifc™), interfaced with a nanospray Flex ion source, and equipped with an Acclaim PepMap RSLC analytical column (25 cm × 75 µm, nanoViper, C18, 2 µm) at a flow rate of 300 nL/min. The gradients used were developed with 0.1% FA/99.9% H₂O (Eluant A) and 0.1% FA/80% acetonitrile/19.9% H₂O (Eluant B): 1–5% B in 2.5 min followed by 5–22% B in 60 min, 22–35% B in 15 min followed by 35–99% B in 7 min. Solvent B was held at 99% until 2.5 min, followed by re-equilibration. In data-dependent acquisition mode, the Orbitrap MS was operated in positive mode with a full scan range of 350–1500 m/z at a resolution of 60,000. To automatic gain control (AGC) was set to “custom” mode, with a normalized AGC target of 200% and a maximum injection time of 100 ms. Precursor ion selection width was set at 2 Da. Peptide fragmentation was triggered by higher-energy collisional dissociation using a collision energy of 26%. Four technical replicates were carried out to maximize the identification of SNO proteins.

The mass spectrometry conditions were as follows: spray voltage, 1.9 kV; capillary temperature, 250 °C; S-lens RF level, 45%; exclusion duration, 45 s. Data acquisition was performed using Xcalibur 2.0 software. MS data processing was performed as described in our previous investigation⁷².

Bioinformatics analysis

For functional enrichment analysis of GO processes and pathways, the differentially SNO proteins were uploaded into the MetaCore software. A Reactome pathway analysis (with a score of greater than 0.7) was conducted using the STRING database (http://string-db.org/). The visualization of protein interactions was conducted utilizing the Cytoscape software. Utilizing the aforementioned software for analysis, those with FDR < 0.1 were regarded as statistically significant.

Immunofluorescence (IF)

Freshly isolated whole brain tissues from mice in each group were immediately immersed in 4% (w/v) paraformaldehyde (PFA) solution and fixed at 4 °C for 24 h. After fixation, the brain tissues were subjected to gradient sucrose cryoprotection, snap-frozen, and embedded in optimal cutting temperature compound. Serial sections with a thickness of 20 μm were prepared using a cryostat. The frozen sections were rinsed 3 times with 1× phosphate-buffered saline (PBS), permeabilized, and blocked with 10% BSA for 1 h at RT. Primary antibodies against SYN1 (1:2000) and PSD-95 (1:2000) were simultaneously applied to the sections and incubated overnight at 4 °C. On the next day, the sections were rinsed 3 times with 1× PBS, followed by incubation with the corresponding fluorophore-conjugated secondary antibodies for 2 h at RT. After 3 additional rinses with 1× PBS, the sections were mounted with an anti-fluorescence quenching mounting medium containing 4’,6-diamidino-2-phenylindole (DAPI)⁷³. Finally, images of the hippocampal region in the brain sections from each group were acquired using a Nikon Eclipse C1 confocal microscope (Nikon, Tokyo, Japan). The acquired images were imported into ZEISS ZEN software (Carl Zeiss, Jena, Germany) for analysis, and the relative fluorescence intensity (RFI) of the target proteins was subsequently quantified.

Hematoxylin & eosin (H&E) staining

Brain tissue samples were fixed in 4% (w/v) paraformaldehyde at 4 °C for 24 h, after which they were routinely paraffin-embedded and sectioned into 5 μm-thick serial sections. The paraffin sections were deparaffinized, rehydrated, and stained with hematoxylin and eosin (H&E) following the manufacturer’s instructions of the commercial H&E staining kit (Servicebio, Wuhan, China). The stained sections were then subjected to gradient dehydration, cleared, and mounted with neutral balsam. Subsequently, the hippocampal region in the brain tissue sections was observed under a bright-field light microscope (Olympus, Tokyo, Japan) to identify histopathological alterations in the target region.

Biochemical indicators testing

In order to detect the levels of SOD and MDA in the brain samples, an appropriate quantity of saline was added to the tissue and homogenized. Subsequently, the mixture was subjected to centrifugation at 1000 × g at 4 °C for 10 min, and the supernatant was extracted, followed by measurement using ELISA assay kits (Shanghai Enzyme Biotechnology Co., Ltd, China) in accordance with the manufacturer’s instructions. Absorbance was read at 450 nm.

To measure whole brain tissue NO levels, 0.1 g of whole brain tissue was homogenized in 0.9 mL of 0.86% saline on ice. The mixture was then centrifuged at 2500 × g for 10 min, and the supernatant was collected, followed by determination of NO levels in the supernatant using a NO assay kit (Nanjing Jiancheng Bioengineering Institute Co., Ltd, China). The absorbance was measured at 550 nm.

To determine the level of 3-NT in cells, following pretreatment of cell samples, the supernatants were collected. The concentration of 3-NT in the supernatants was measured using the ElaBoX™ 3-NT ELISA Kit (Solarbio Science & Technology Co., Ltd, China). The absorbance was detected at a wavelength of 450 nm.

For the detection of cellular 4-HNE levels, cell samples were pretreated, followed by collection of the cell culture supernatants. The concentration of 4-HNE in the supernatants was determined using a commercial 4-HNE ELISA Kit (Shanghai Youxuan Biological Technology Co., Ltd, China). Absorbance was measured at a wavelength of 450 nm.

Western blot (WB) analysis

To detect the expression levels of p47^phox, P-p47^phox, pan-PKC, p67^phox, RAC2, NOX1, NOX5, nNOS, SYN1, and PSD-95, the tissues were homogenized in radioimmunoprecipitation assay lysis buffer (1:5, w/v) containing phenylmethylsulfonyl fluoride and a protease inhibitor cocktail (100:1:1, v/v/v) at ice-cold temperature. The mixture was then centrifuged at 12,000 × g for 15 min at 4 °C. The bicinchoninic acid assay was employed to determine protein concentration. A 10% SDS-PAGE gel was prepared using an Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) gel kit, and 20 μg of sample was loaded into each well for electrophoresis and subsequent membrane transfer. Primary antibodies p47^phox (1:1000), P-p47^phox (1:1000), pan-PKC (1:5000), p67^phox (1:2000), RAC2 (1:8000), NOX1 (1:1000), NOX5 (1:800), nNOS (1:1000), SYN1 (1:5000), PSD-95 (1:8000) and internal reference GAPDH (1:1000) were diluted to the indicated ratios with their respective antibody diluents. The transferred samples were then blocked at RT with 5% (w/v) skimmed milk powder in Tris-buffered saline containing 0.05% Tween 20 (TBST) for 1 h, followed by overnight incubation with primary antibodies at 4 °C. Following five washes with TBST, horseradish peroxidase-conjugated secondary antibodies were incubated with the membranes for 1 h at RT with gentle shaking. Protein bands were visualized using enhanced chemiluminescence (ECL, Beyotime Biotechnology Co., Ltd, Shanghai, China) solution and imaged with a ChemiDoc MP imaging system (Bio-Rad, Hercules, CA, USA). The quantification of the protein bands was performed using ImageJ software.

Quantitative real-time polymerase chain reaction (qRT-PCR) analysis

RNA was extracted from the hippocampal tissues using Trizol reagent (Tiangen Biotechnology Co., Ltd, Beijing, China), and the quality of the RNA was monitored using a NanoDrop One 5225 UV-visible spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). A commercial reverse transcriptase kit (Tiangen Biotechnology Co., Ltd, Beijing, China) was employed for cDNA synthesis. The qRT-PCR analysis was conducted using the SYBR Green system from qTOWER3G (Analytik Jena AG, Germany). Primers were purchased from Sangon Biotech (Shanghai, China). The primer sequences are shown below: SYN1 (F: 5′-AGC TCA ACA AAT CCC AGT CTC T-3′; R: 5′-CGG ATG GTC TCA GCT TTC AC-3’); PSD-95 (F: 5′-TGA GAT CAG TCA TAG CAG CTA CT-3′; R: 5′-CTT CCT CCC CTA GCA GGT CC-3′). GAPDH (F: 5′-GCC AAG GTC ATC CAT GAC AAC-3′; R: 5′-AGT GTA GCC CAG GAT GCC C-3′). The relative gene expression levels were calculated using the 2^-ΔΔCt method, with GAPDH as the housekeeping gene.

Cell culture

PC12 cells were cultured in 10% RPMI 1640 medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin, in a humidified 5% CO₂ incubator at 37 °C. The culture medium was refreshed every 2–3 days, and cells were passaged with 0.25% trypsin-EDTA when they reached 90% confluence.

Cell viability

Cell proliferation and viability were evaluated using the CCK-8 assay. Different concentrations of H₂O₂ were employed to determine the half-maximal inhibitory concentration (IC₅₀). Cells were adjusted to a density of 2 × 10⁴ cells per well and seeded into 96-well plates. For the OSBG treatment group, cells were co-incubated with OSBG and H₂O₂ for 24 h. After induction, the supernatant was discarded, and 100 μL of fresh medium was added to each well. Subsequently, 10 μL of CCK-8 reagent was added to each well, followed by incubation at 37 °C in a 5% CO₂ atmosphere for 1 h. The optical density (OD) was measured at a wavelength of 450 nm using a Synergy4 microplate reader (BioTek, Winooski, VT, USA).

ROS determination

Briefly, following cell harvest, 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) was diluted at a ratio of 1:1000 in serum-free medium (SFM) to achieve a final concentration of 10 μM, and the probe was then loaded onto cells. Cells seeded in 6-well plates were harvested and resuspended in an appropriate volume of diluted DCFH-DA at a density of 1 × 10⁶–2 × 10⁷ cells/mL, followed by incubation at 37 °C. During incubation, the cell suspension was gently inverted every 5 min for a total of 20 min to ensure sufficient contact between cells and the probe. Subsequently, cells were washed with SFM to thoroughly remove residual DCFH-DA. The fluorescence intensity of intracellular ROS was acquired using a Leica DMi8 inverted fluorescence microscope (Wetzlar, Germany). Fluorescence intensity was quantified using LAS X software (Wetzlar, Germany).

Biotin-switch assay

Biotin-switch assays were performed as described previously⁷⁴. Briefly, brain tissues were lysed in HEN buffer containing 1% Triton X-100 (250 mM HEPES, 1 mM EDTA, 0.1 mM neocuproine). Free thiols were blocked with methyl-methanethiosulfonate. Proteins were precipitated with acetone and resuspended in HEN buffer supplemented with 1% SDS. Nitrosothiols were selectively reduced with ascorbic acid to restore thiol groups, which were subsequently biotinylated with 1 mM biotin-HPDP. Biotinylated proteins were precipitated using streptavidin-agarose beads and analyzed by Western blotting. In control experiments, ascorbic acid was omitted to verify the specificity of the observed biotinylated bands.

Statistical analysis

All data were expressed as the mean ± standard error of the mean (SEM). One-way analysis of variance (ANOVA) was performed using IBM SPSS Statistics 25.0, and post hoc multiple comparisons were performed using the Dunnett-t method. In terms of significance analysis, p < 0.05 was thought to be significant.

Author contributions

P.Y., J.L., and W.X. designed experiments and data analyses; L.P., Y.L., and S.S. performed the experimental work; Y.W., P.Y., and J.L. analyzed data and wrote the manuscript; Z.Q. revised the manuscript. H.Y. supervised the project. All authors reviewed the manuscript.

Data availability

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41538-026-00812-9.